Readiness quiz week five Stats 7000

Readiness quiz week five Stats 7000 1. If the risk of thromboembolism is 10% in the treatment group and 15% in the control group, the absolute risk reduction is a. 10% b. 5% c. 50% d. 33% 2. If the risk of thromboembolism is 10% in the treatment group and 15% in the control group, the relative risk reduction is a. 10% b. 5% c. 50% d. 33% 3. Suppose the attributable risk (aka risk difference) in tumor progression between chemotherapy and radiation therapy for kidney cancer is 3% (radiation minus chemo). The number needed to treat in this scenario is a. 3 b. 33 c. 97 d. 1 4. Suppose the risk of melanoma for 18-35 year olds is 18%, while in those 65+ years of age, it is 54%. The relative risk of melanoma for 18-35-year-olds compared to those 65+ is a. 3 b. 0.33 c. 36% d. 50% 5. Suppose the risk of diabetes for adult Hispanics/Latinos is 12%, while for non-Hispanic/Latinos it is 42%. The relative risk of melanoma for Non-Hispanic/Latinos compared to Hispanic/Latinos is a. 3.5 b. 0.286 c. 30% d. -30% 6. The key assumptions required to appropriately interpret the results of a prospective or experimental study include (choose all that apply) a. Random sample b. Independent observations c. Similarity of groups being compared d. Sufficiently large sample size e. Sufficiently narrow confidence interval 7. A single relative risk can be computed to compare the ‘White,’ ‘Black,’ and ‘Other’ race groups for access to emergency department care (Yes, have access or No do not) simultaneously. a. True b. False 8. An investigator conducted a study and concluded that the relative risk of brain cancer for those living near electric power lines compared to a control sample that does not near live near such power lines is 2.7. This number is best described as a. Valid b. Invalid c. Accurate based on the available study data d. Inconclusive 9. Odds ratios are valid to compute in case-control studies a. True b. False

10. Relative risk (aka risk ratios) are valid to compute in case-control studies. a. True b. False 11. Suppose an odds ratio comparing heart disease for females to males was computed as 0.74. This can be interpreted as (choose the best answer) a. Females are 26% less likely than males to have heart disease b. Males are 26% more likely than females to have heart disease c. Females are 74% less likely than males to have heart disease d. Males are 0.74% as likely as females to have heart disease 12. Valid test(s) to use to test for a difference in proportions include (choose all that apply): a. Fischer’s exact test b. Chi-square test c. Paired t-test d. Normality test 13. For virtually all 95% confidence intervals for odds ratios that do not include/contain one, the p-value for the corresponding Fisher’s exact test will be <0.05. a. True b. False 14. In case-control studies, when determining/selecting matched controls, matching not only for variables but also for the exposure of interest is desirable. a. True b. False 15. Cases, in case-control studies, are defined as all the people in the population of interest with the disease or condition of interest that develops during the defined period of time. a. True b. False 16. In most case-control studies, cases are defined as those who already have the disease/condition of interest, not those who develop it. a. True b. False 17. Case-control studies may be either retrospective or prospective. a. True b. False 18. Selection bias may occur in case-control studies when the selected controls are not truly representative of the population that produced the cases. a. True b. False 19. In case-control studies, it is crucial to ensure that the same data that is collected on the cases is available and collected on the controls in order to ensure a fair comparison of the groups. a. True b. False 20. Suppose a researcher was interested in generating a narrower confidence interval for the difference between the mean final exam score between those taught in a traditional (lecture) format and those taught in a ‘flipped’ classroom setting. The best way (i.e., most realistic and effective way to have informative results) to achieve a narrower interval is to a. Reduce variability b. Increase sample size c. Increase the degree of confidence d. Decrease the mean difference 21. For all the hypothesis tests we covered in this course to this point, it is possible for an investigator to essentially ‘buy’ themselves a statistically significant result by spending additional money to increase the sample size of their study.

d. A normality test 32. An investigator sought to determine whether the mean weight following an eight week exercise and diet program had been reduced. The best statistical test (among the available choices) to determine this is a. A paired t-test b. An unpaired t-test c. A chi-square test d. Fisher’s exact test 33. A paired t-test looks at the differences in measurements between two matched subjects or measurements made before and after and experimental intervention. a. True b. False 34. While an unpaired t-test looks at the difference in means between two groups, the paired t-test looks at the mean of the differences between individual matched observations. a. True b. False 35. McNemar’s test is utilized for paired data when the outcome is binary. a. True b. False 36. When conducting a paired t-test, the differences should always be computed such that they are all positive in order to ensure consistency across all the pairs. a. True b. False